Concept of Elasticsearch cluster

pyerunka · October 31, 2019, 5:45am

Thanks for reply!!!
One humble request, Can you please hide ip address used in earlier comments from your side?
And please do not use it directly. You can mention ip_address_of_master.

Master node is working. I am getting below responses:

curl -X GET ip_of_master:9200
{
  "name" : "Master-node",
  "cluster_name" : "cluster",
  "cluster_uuid" : "yQvifHj-Tw2RNMCItAT7SA",
  "version" : {
    "number" : "7.2.0",
    "build_flavor" : "default",
    "build_type" : "zip",
    "build_hash" : "508c38a",
    "build_date" : "2019-06-20T15:54:18.811730Z",
    "build_snapshot" : false,
    "lucene_version" : "8.0.0",
    "minimum_wire_compatibility_version" : "6.8.0",
    "minimum_index_compatibility_version" : "6.0.0-beta1"
  },
  "tagline" : "You Know, for Search"
} 

curl -X GET ip_of_master:9300
curl: (7) Failed to connect to "ip_of_master" port 9300: Timed out

Thanks,
Priyanka

Dominic_Page · October 31, 2019, 9:28am

Hi @pyerunka

No problem, I've redacted the IPs in my previous posts

From these 2 calls:

curl -X GET XXX.XXX.XXX.XXX:9200

indicates that the data node has visibility of the master node's HTTP Port 9200, however, to form a cluster it must be able to access to the master node's Transport Port 9300

curl -X GET XXX.XXX.XXX.XXX:9300

is not yet connecting, so try that again with the verbose flag, you would hope to see a conversation something like this:

$ curl -vX GET IP_OF_MASTER_NODE:9300
* About to connect() to IP_OF_MASTER_NODE port 9300 (#0) 
*   Trying XXX.XXX.XXX.XXX...
* Connected to server1 (XXX.XXX.XXX.XXX) port 9300 (#0)
> GET / HTTP/1.1
> User-Agent: curl/7.29.0
> Host: IP_OF_MASTER_NODE:9300
> Accept: */*
>
* Connection #0 to host server1 left intact
This is not an HTTP port

Dominic_Page · November 11, 2019, 6:01pm

Hi @pyerunka

Were you able to obtain additional information using

$ curl -vX GET IP_OF_MASTER_NODE:9300

from the Data node?

pyerunka · November 12, 2019, 5:29am

Hello @Dominic_Page,

I tried using above command on data node. I am getting below response:

* Rebuilt URL to: IP_OF_MASTER_NODE:9300/
* Hostname was NOT found in DNS cache
*   Trying IP_OF_MASTER_NODE...
* connect to IP_OF_MASTER_NODE port 9300 failed: Timed out
* Failed to connect to IP_OF_MASTER_NODE port 9300: Timed out
* Closing connection 0
curl: (7) Failed to connect to IP_OF_MASTER_NODE port 9300: Timed out

Regards,
Priyanka

Dominic_Page · November 12, 2019, 9:22am

Hi @pyerunka

Just checking - did you change IP_OF_MASTER_NODE to the IP of your master node?

pyerunka · November 12, 2019, 9:28am

Hello @Dominic_Page,

Yes, I have changed it to actual IP address and then checked.
While giving reply, I have mentioned "IP_OF_MASTER_NODE" only.

Regards,
Priyanka

Dominic_Page · November 12, 2019, 9:51am

Hi @pyerunka

It appears there is still some network issue blocking communications between the servers.

For a healthy cluster, it is a good idea to reduce the number of network hops between servers.

If you run

$ traceroute IP_OF_MASTER_NODE

from the data node: how many network hops are shown?

You should see an output something like this:

$ traceroute IP_OF_MASTER_NODE
traceroute to IP_OF_MASTER_NODE (xxx.xxx.xx.xx), xx hops max, xx byte packets
 1  xx.xxx.xxx.x (xx.xxx.xxx.x)  x.xxx ms  x.xxx ms  x.xxx ms
 3  xxx.xx.xxx.xxx (xxx.xx.xxx.xxx)  x.xxx ms  x.xxx ms  x.xxx ms
 4  xxx.xx.xxx.xxx (xxx.xx.xxx.xxx)  x.xxx ms  x.xxx ms  x.xxx ms
 5  gix-x-x-x-xx-crx.mad.router.colt.net (xxx.xx.xx.xxx)  xx.xxx ms  x.xxx ms  x.xxx ms
 6  xex-x-x-prx.mad.router.colt.net (xxx.xx.xx.xxx)  x.xxx ms
    xex-x-x-prx.mad.router.colt.net (xxx.xx.xx.xxx)  x.xxx ms  x.xxx ms
 7  xex-x-x-prx.mad.router.colt.net (xxx.xx.xx.xxx)  x.xxx ms

pyerunka · November 13, 2019, 4:42am

Hello @Dominic_Page,

I have raised this network issue which is blocking communications between two servers to our network team.

And i also tried $ traceroute IP_OF_MASTER_NODE from the data node with actual ip address of master node. It is showing me below error:

 'traceroute' is not recognized as an internal or external command,
operable program or batch file.

Regards,
Priyanka

MiTschMR · November 13, 2019, 6:54am

Hi @pyerunka

You sure know about traceroute, don't you? The error message shows that you are using Windows as your operating system. If traceroute is not accepted there, then a quick search indicates that the correct command on Windows is tracert.

How to Use TRACERT to Troubleshoot TCP/IP Problems in Windows

Usage of tracert:

Usage: tracert [-d] [-h maximum_hops] [-j host-list] [-w timeout]
               [-R] [-S srcaddr] [-4] [-6] target_name

Options:
    -d                 Do not resolve addresses to hostnames.
    -h maximum_hops    Maximum number of hops to search for target.
    -j host-list       Loose source route along host-list (IPv4-only).
    -w timeout         Wait timeout milliseconds for each reply.
    -R                 Trace round-trip path (IPv6-only).
    -S srcaddr         Source address to use (IPv6-only).
    -4                 Force using IPv4.
    -6                 Force using IPv6.

pyerunka · November 13, 2019, 7:00am

Hello @MiTschMR,

Thanks for your reply!!!
I do not know about this command actually.
and when i have tried tracert command, i did not get any quick search indication for correct command.
I will refer above help.

Thanks,
Priyanka

pyerunka · November 18, 2019, 8:30am

Hello @Dominic_Page,

Ports issue has been resolved now.
I am able to connect to another server to form a cluster using ES.
How to check that which cluster nodes are running on my ES?
and after that when i have tried to start kibana, it is not working for me.

Can you guide me more on the same.

Regards,
Priyanka

Dominic_Page · November 18, 2019, 1:34pm

HI @pyerunka

Great news that your port issue is resolved!

How to check that which cluster nodes are running on my ES?

Try

$curl YOUR_IP:9200/_cat/nodes?v\&h=ip,node.role,master,name

This will list the nodes which are currently members of the cluster

Re Kibana:

and after that when i have tried to start kibana, it is not working for me.

Can you open a new discussion for this question?

BR - Dominic

pyerunka · November 19, 2019, 7:17am

Hello @Dominic_Page,

Thanks for your help!!!
I have open new discussion for kibana issue.
For Es cluster, I can see one head plugin we can download.
I am trying to download it using "plugin --install mobz/elasticsearch-head" . but not able to create.
Can you help me this?

Regards,
Priyanka

Dominic_Page · November 19, 2019, 7:22am

Hi @pyerunka

The elasticsearch-head plugin is outside my area of expertise

I believe you can find the docs here: https://github.com/mobz/elasticsearch-head

BR,

Dominic

Dominic_Page · November 19, 2019, 7:24am

Hi again @pyerunka

One update, I took a quick look at that, I think you have to run as a standalone server for Elasticsearch >= 5.0 - the guide is here

Hope this helps,

Dominic

system · December 17, 2019, 7:24am

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.